Subhalos without progenitor

Fabio Lesjak
  • 2 Jul '19

Hi Dylan,

I am analysing a particular set of 116 subhalos at z=0, and I noticed that over half of them (64 to be exact) have no progenitor and can therefore not be traced back in time.
In an earlier post you wrote that this might happen for very small subhalos or complex mergers. The stellar masses of all of these subhalo are between 10^8.5 and 10^10 solar masses, so I would assume that this can not be the problem for all of them.

Do you have any idea how this could happen otherwise, or what the best way would be to further analyse these subhalos to gain some insights about what is going on with them?

Here is the list of subhalo ids of these 64 subhalos (in snapshot 99):
[343, 521, 616, 912, 17326, 31619, 41945, 42035, 42242, 52846, 52998, 69863, 76217, 76291, 76309, 76314, 76320, 76408, 76495, 108082, 108125, 114424, 114429, 114457, 114493, 114635, 172764, 175325, 179979, 192940, 192980, 192986, 192991, 197143, 197172, 202688, 220595, 228729, 231778, 237991, 237999, 260912, 260928, 260930, 266585, 290668, 294525, 294527, 294535, 294549, 298221, 298240, 326243, 333702, 361193, 372788, 382930, 382932, 382940, 383957, 392859, 410576, 412343, 485544]

Thanks in advance!

Dylan Nelson
  • 2 Jul '19

Hi Fabio,

I would certainly be cautious interpreting anything about these - if a subhalo has no merger tree, clearly it has had a strange history. 343 for instance looks like a sub-component of a satellite (its SubhaloParent is 2, not 0). I suspect many of these objects are subhalos which should have SubhaloFlag==0, i.e. should be viewed with caution. If you really want to track them, then you can just load their DM or stellar IDs, and crossmatch to the IDs in the previous snapshot, to figure out where they are (e.g. their Coordinates evolution).

Fabio Lesjak
  • 3 Jul '19

Hi Dylan,

I actually applied a filter concerning the SubhaloFlag prior to my selection, so all of these subhalos have SubhaloFlag==1. Thanks for the idea to track them back by manually crossmatching, I am going to try that approach and see if I gain any new insights.

  • Page 1 of 1